GitLab

This patch rewrites the detection logic for composited animations that
have neutral keyframes.

The previous code was relatively expensive and failed a DCHECK when
custom properties were used. This patch moves the use counting down into
where we actually handle compositing neutral keyframes.

BUG=679994

Review-Url: https://codereview.chromium.org/2620303002
Cr-Commit-Position: refs/heads/master@{#443180}

and it is false.

It was confusing that the flag seemed to check if enable some
specific members (attributes, methods, and constants), but it
should check everything of the interface.
It means if the flag is false, we don't have to check all
members, e.g. indexed properties and iterators.

BUG=None

Review-Url: https://codereview.chromium.org/2620153004
Cr-Commit-Position: refs/heads/master@{#443179}

[layout-tests] Anticipate a manual rebaseline for imported/wpt/html/browsers/origin/cross-origin-objects/cross-origin-objects-exceptions.html

R=jochen@chromium.org
BUG=chromium:575167

Review-Url: https://codereview.chromium.org/2621373002
Cr-Commit-Position: refs/heads/master@{#443178}

BUG=670617
TEST=Show the on-screen keyboard on non-touchable display and
     switch to another IME; then check that the on-screen keyboard
     remains on the same display.

Review-Url: https://codereview.chromium.org/2546963002
Cr-Commit-Position: refs/heads/master@{#443177}

[Presentation API] (1-UA) Split PresentationServiceDelegateImpl(PSDImpl) into ControllerPSDImpl and ReceiverPSDImpl

(Splitting of https://codereview.chromium.org/2379703002/)

In an offscreen presentation, we have controlling frame (page that starts a PresentationRequest with url) and receiver frame (page corresponding to the url, rendered as offscreen tab). They talk to MR extension and OffscreenPresentationManager (OPM) in different ways. It makes sense to split them.

- create ReceiverPSDImpl class
- PresentationServiceImpl (PSImpl) class takes controller_delegate_ and receiver_delegate_ objects. For controlling frame, receiver_delegate_ = nullptr; For receiver frame controller_delegate_ = nullptr
- Added PSImpl::SetPresentationConnection() - will be called in PresentationDispatcher::OnSessionCreated(), PresentationDispatcher::OnReceiverConnectionAvailable(); will call OPM::RegisterOffscreenPresentationController()
- Added PSImpl::OnReceiverConnectionAvailable() - will be registered to OPM as receiver callback

PSImpl.receiver_delegate_ is always nullptr now (make sure things wont break). It will be set in offscreen_tab.cc in next CL.

More details in code comments and design doc (use chromium.org account): https://docs.google.com/document/d/1XM3jhMJTQyhEC5PDAAJFNIaKh6UUEihqZDz_ztEe4Co/edit#heading=h.hadpx5oi0gml

BUG=525660

Review-Url: https://codereview.chromium.org/2477573002
Cr-Commit-Position: refs/heads/master@{#443176}

RenderThreadImpl::Shutdown has been shutting down Blink and V8 gracefully,
but the graceful shutdown has caused tons of use-after-free bugs
(and many engineers has spent lots of time fixing ordering issues around the shutdown).

As discussed in chromium-dev@ (https://groups.google.com/a/chromium.org/d/topic/chromium-dev/OLS4JSZvowI/discussion) and
platform-architecture-dev@ (https://groups.google.com/a/chromium.org/d/topic/platform-architecture-dev/Zc12k91NTFk/discussion),
there is no reason we have to shut down the renderer gracefully.
It's just causing use-after-free bugs and wasting performance.

Hence this CL removes RenderThreadImpl::Shutdown and instead calls _exit(0) in ~ChildProcess.
We need a couple of special handling in a single-process mode. See the comment in ~ChildProcess
for more details.

BUG=639244

Review-Url: https://codereview.chromium.org/2309153002
Cr-Commit-Position: refs/heads/master@{#443175}

The reland adds a fix for a crash in window animation code when restoring
a session that has a minimized window. The animation tries to compute a
shelf item position before the shelf is initialized. The animation is not
visible to the user. Returning an empty rect results in a default target
position, which is fine in this case.

Added guards for access to ShelfView before shelf is initialized.

Original CL description:

Previously chrome --mash would always create the shelf on startup, even at
the login screen. Now it waits until login is complete, like classic ash.

Ash watches for SessionState::ACTIVE via a SessionStateObserver to create the
shelf, rather than NOTIFICATION_LOGIN_USER_PROFILE_PREPARED. For most login
flows this doesn't matter, but for supervised user creation it means that the
shelf is not created until the flow is completed. This is an improvement
because in the old code the shelf would be created too early and had to be
explicitly hidden.

Always create a SessionControllerClient in chrome, even in classic ash. This
allows classic ash to use the mojo pathway via ash::SessionController.

Remove WmShell::ShowShelf(), which was introduced in crrev.com/10693003 to
delay showing the shelf until post-login OOBE (like avatar picture select)
was complete. It does not seem to be needed anymore, either in production or
in tests.

Change AshTestImplMus to simulate a user logging in which is required to
create the shelf on the primary display and also to have a non-zero user
count to create the shelf on additional displays.

Remove some unnecessary OS_CHROMEOS ifdefs in //c/b/ui/ash.

BUG=666021, 679513
TEST=ash_unittests and chrome browser_tests

Committed: https://crrev.com/7f99e933e595824281102ff6737075dccbbc4d5c
Cr-Commit-Position: refs/heads/master@{#439845}

Reverted:

Review-Url: https://codereview.chromium.org/2619943002
Cr-Original-Original-Commit-Position: refs/heads/master@{#442484}
Review-Url: https://codereview.chromium.org/2625733003
Cr-Original-Commit-Position: refs/heads/master@{#443106}
Committed: https://chromium.googlesource.com/chromium/src/+/75298de72cf7c25e137c3c5e63e4d01aa0569c1c
Review-Url: https://codereview.chromium.org/2625733003
Cr-Commit-Position: refs/heads/master@{#443174}

The mojo responses can be reached to the callback even after the PendingRequest
is removed. That didn't happen on the legacy IPC because the receiver object was
owned by PendingRequest. This patch is to allow the situation that the
PendingRequest is null when the response has received.

BUG=676984

Review-Url: https://codereview.chromium.org/2628553002
Cr-Commit-Position: refs/heads/master@{#443173}

https://crrev.com/2627543002 removes the non-mojo path. Let's remove the
non-mojo tests and the flag to run non-mojo path.

BUG=629701

Review-Url: https://codereview.chromium.org/2619883004
Cr-Commit-Position: refs/heads/master@{#443172}

The metadata stands for the pixel size of the image or the video.
Dimension is more appropriate name for this than resolution.

TEST=manually.
BUG=680011
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2622393003
Cr-Commit-Position: refs/heads/master@{#443171}

This also lets the FilteredListWidget be smaller if it has few results.

BUG=662081

Review-Url: https://codereview.chromium.org/2599623002
Cr-Commit-Position: refs/heads/master@{#443170}

Updated the open file button's outline to be consistent with the
outline of metadata box toggle button.

TEST=manually.
BUG=663669
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2623413002
Cr-Commit-Position: refs/heads/master@{#443169}

https://pdfium.googlesource.com/pdfium.git/+log/d805eec52f6a..e8b0281d1342

$ git log d805eec52..e8b0281d1 --date=short --no-merges --format='%ad %ae %s'
2017-01-11 dsinclair Disable CQ bot retries

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2632443002
Cr-Commit-Position: refs/heads/master@{#443168}

A ResourceLoader is held as a raw-ptr in WebURLLoaderImpl and hence it must
cancel the WebURLLoaderImpl when it's destroyed. We thought
blink::Resource::allClientsAndObserversRemoved did the job but it's wrong:
the cancel timer doesn't work when the resource is about to be destroyed.

BUG=675947
R=hiroshige@chromium.org, japhet@chromium.org

Review-Url: https://codereview.chromium.org/2625823003
Cr-Commit-Position: refs/heads/master@{#443167}

As part of this I made sure the definitions and declarations matched
up. This resulted in reordering a couple of things. In addition the
only other non-move I did was moving what was CreateLayoutManagers()
into InitLayoutManagers() and CloseChildWindowsImpl() into
CloseChildWindows().

BUG=671246
TEST=none
R=jamescook@chromium.org

Review-Url: https://codereview.chromium.org/2628973002
Cr-Original-Commit-Position: refs/heads/master@{#443109}
Committed: https://chromium.googlesource.com/chromium/src/+/d20a89e7b56ad53a58f19edf4775528d066577d5
Review-Url: https://codereview.chromium.org/2628973002
Cr-Commit-Position: refs/heads/master@{#443166}

This CL adds a static method to VTR to query which
one, if any, is the preferred video codec.

A new singleton Lazy Leaky class CodecEnumerator is
added to encapsulate poking the VEA to see which
codecs are supported, and to further query
a) the preferred codec id
b) if a given codec is supported
and wraps the previous CodecIdToVEAProfile()
functionality.

BUG=679946

Review-Url: https://codereview.chromium.org/2624053002
Cr-Commit-Position: refs/heads/master@{#443165}

https://chromium.googlesource.com/chromium/src/+/13a86c2faeebc

BUG=453414
TBR=ktyliu@chromium.org

Review-Url: https://codereview.chromium.org/2628173002 .
Cr-Commit-Position: refs/heads/master@{#443164}

This CL also removes a skip from W3CImportExpectations that should no
longer be necessary.

BUG=676491

Review-Url: https://codereview.chromium.org/2628913003
Cr-Commit-Position: refs/heads/master@{#443163}

Everything necessary to use the existing display to touch input-device
association/transform code is accessible from mus now.
ScreenManagerOzone creates a TouchTransformController and tells it to
compute new transforms when the display configuration changes. That's
it.

Delete existing touch code that was added to the WS to associate a
single display and touch input-device.

BUG=679747

Review-Url: https://codereview.chromium.org/2627623003
Cr-Commit-Position: refs/heads/master@{#443162}

This change allows geolocation access for the default search engine if
it would have access to geolocation information via the X-Geo header.

To effect this, a new setting is enabled controlling whether the default
search engine has geolocation access. This is only used if the search
engine supports the X-Geo header, i.e. it is Google.

The setting is consulted for the DSE both when deciding if the X-Geo
header should be added, and when deciding if geolocation API access
should be granted.

To allow the UI to make sense and provide the greatest clarity, the
setting should always be consistent with the underlying content setting
for the DSE's origin. In this end it is initialized accordingly and
various checks are in place to keep it consistent.

BUG=674389

Review-Url: https://codereview.chromium.org/2612993002
Cr-Commit-Position: refs/heads/master@{#443161}

https://skia.googlesource.com/skia.git/+log/8d80bb5f20d8..e4ca362259a3

$ git log 8d80bb5f2..e4ca36225 --date=short --no-merges --format='%ad %ae %s'
2017-01-11 reed add win subclass of raster-allocator

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
TBR=kjlubick@google.com

Review-Url: https://codereview.chromium.org/2625933005
Cr-Commit-Position: refs/heads/master@{#443160}

Now if Map() failed, we clear |data_| and |size_|, but |pp_resource_| is
not cleared. Therefore, if the caller calls buffer.is_null() it will
return false. Since the Init() failed, buffer.is_null() should return
true.

In this CL, if Map() failed, we call Clear() to explicitly clear
|pp_resource_|.

BUG=679801

Review-Url: https://codereview.chromium.org/2629543002
Cr-Commit-Position: refs/heads/master@{#443159}

Purpose: This should prevent upstream changes in manifest from breaking
import or test runner unexpectedly.

BUG=678077

Review-Url: https://codereview.chromium.org/2630493002
Cr-Commit-Position: refs/heads/master@{#443158}

Specifically, a 2.55 multiplication factor should be used, and
serialization should be to two or three digits according to section
<alphavalue> in https://drafts.csswg.org/cssom/#serializing-css-values

This fixes test case "background-color: rgba(5, 7, 10, 0.5)" in
https://w3c-test.org/cssom/serialize-values.html and also that the
implementation is now same as Firefox as noted in the bug.

Note a large number of tests need to be fixed, many of them because:
- *-expected.txt specifically had bad alpha values such as 0.501961 or
- an alpha value of 0.5 was previously 127 but should have been 128.
The latter applies to the updated expected images too where the only
difference was tiny bit of alpha (and rgb when alpha blended) change.

BUG=453414

Review-Url: https://codereview.chromium.org/2608423002
Cr-Commit-Position: refs/heads/master@{#443157}

Previously, when tab button was clicked up to thumbnail view option in
gallery, the tooltip for the option did not work. This is also true of
slide view option in thumbnail mode. Both issues are resolved with this
patch.

BUG=549547
TEST=manually tested in the issue
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2626973004
Cr-Commit-Position: refs/heads/master@{#443156}

BUG=405845

Review-Url: https://codereview.chromium.org/2606733003
Cr-Commit-Position: refs/heads/master@{#443155}

Changes:
stylus 'touch' event is no longer mouse event based on the cl: https://codereview.chromium.org/2507503002/, which leads to this cleanup change.

BUG=633304
TEST=tested that stylus touch event will not wake the device (both in
laptop mode and tablet mode) on kevin.

Review-Url: https://codereview.chromium.org/2621373003
Cr-Commit-Position: refs/heads/master@{#443154}

BUG=None

Review-Url: https://codereview.chromium.org/2543643006
Cr-Commit-Position: refs/heads/master@{#443153}

The interface is no longer needed since we've added GetService<T>()
to the manager class.

BUG=672840
TEST=try

Review-Url: https://codereview.chromium.org/2623273003
Cr-Commit-Position: refs/heads/master@{#443152}

It's been implemented as just "user-select" since Chrome 54.

This was accomplished with:

for type in none text auto; do
  git gs "\-webkit-user-select: $type;" -l | \
      xargs perl -p -i -e "s/-webkit-user-select: $type;/user-select: $type;/g"
done

And a bunch of manual alpha sorting of rules :(.

R=fukino@chromium.org
BUG=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2621993006
Cr-Commit-Position: refs/heads/master@{#443151}

On Chromecast we want media logging to be enabled on release builds by
default so that it gets recorded into device logs and sent with
feedback reports from end-users.

Review-Url: https://codereview.chromium.org/2625623002
Cr-Commit-Position: refs/heads/master@{#443150}

Cr-Commit-Position: refs/heads/master@{#443149}

Previous, Gonzales was cherrypicking files from Common module for the
TextRange class which was used as a struct.

BUG=none

Review-Url: https://codereview.chromium.org/2625163002
Cr-Commit-Position: refs/heads/master@{#443148}

BUG=676829
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2629513002
Cr-Commit-Position: refs/heads/master@{#443147}

The main change is static_cast-ing a few chars in ported code to
unsigned chars because on some platforms by default char is signed,
on others it's unsigned.

GitHub page showing the diff:
https://github.com/google/cld3/commit/ae02d6b8a2af41e87c956c7c7d3f651a8b7b9e79

BUG=

Review-Url: https://codereview.chromium.org/2627143002
Cr-Commit-Position: refs/heads/master@{#443146}

BUG=

Review-Url: https://codereview.chromium.org/2628613002
Cr-Commit-Position: refs/heads/master@{#443145}

Review-Url: https://codereview.chromium.org/2604223002
Cr-Commit-Position: refs/heads/master@{#443144}

BUG=none

Review-Url: https://codereview.chromium.org/2626173002
Cr-Commit-Position: refs/heads/master@{#443143}

BUG=657825
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2528513003
Cr-Commit-Position: refs/heads/master@{#443142}

If a mousedown event was set to preventDefault() and the focus was
in a text input field, it won't blur the focus from the text input
field, and the following mouseup event will update the TextInputState of
the widget with the same TextInputType and ShowIme::IF_NEEDED.

If this is a minimize button of an app for example, the widget will hide,
but the later received IPC msg to UpdateTextInputState() will cause the
virtual keyboard to show up.

We need to make sure that the virtual keyboard is not shown as a result of
UpdateTextInputState() unless there's a valid text input client.

BUG=430026
TEST=Enable the virtual keyboard from settings. Open feedback app
(Alt+Shift+i) and minimize it. The virtual keyboard should not show.

Review-Url: https://codereview.chromium.org/2626903004
Cr-Commit-Position: refs/heads/master@{#443141}